GRAVITY MEASUREMENTS. 
51 
Junction, in the bottom of great eroded valleys, and Pikes 
Peak station, on the summit of an isolated mountain. The 
residuals are not proportional to the elevation of the station 
itself above sea level, as is at once seen by comparing|Mount 
Hamilton, Salt Lake, Green River, Grand Junction, and 
Denver, which are of nearly equal elevation. Finally, with 
either system of reduction the residuals point to the possi¬ 
bility that gravity is large on the sea coast as compared with 
the interior. Whether this can have any connection with 
the effects of erosion and deposition of continental matter is 
a question upon which the observations in this series are far 
too limited to throw any clear light, but for investigating 
which certain portions of the United States would furnish 
an excellent field. 
The results of this series would therefore seem to lead to 
the conclusion that general continental elevations are com¬ 
pensated by a deficiency of density in the matter below sea- 
level, but that local topographical irregularities, whether 
elevations or depressions, are not compensated for, but are 
maintained by the partial rigidity of the earth’s crust. The 
residuals with Bouguer’s reduction should then be inter¬ 
preted as a measure of the general deficiency of density, and, 
on the other hand, the residuals with the reduction for eleva¬ 
tion only should be taken as a measure of the lack of local 
compensation, after allowing for uncertainties of observation 
and the effect of local geological conditions. 
Without desiring to advocate any theory as to the condi¬ 
tions of the earth’s crust, it is of interest to note in compari¬ 
son the results for gravity that would be obtained on an ice¬ 
berg floating in the ocean and having an ideal cross-section 
such as that shown in Fig. 1. Here the excess of matter 
